Indian Liver Patient Records with 583 Clinical Observations

583 patient records collected from the North East of Andhra Pradesh, India, used to evaluate prediction algorithms for liver disease. The dataset contains 416 liver patient records and 167 non-liver patient records, with 441 records from male patients and 142 from female patients. It was sourced from the UCI Machine Learning Repository.

Use Cases

• Train a binary classifier to predict liver disease based on patient age, gender, and bilirubin levels.
• Analyze the relationship between liver enzymes like Alkaline Phosphotase and Alamine Aminotransferase and disease status.
• Investigate potential demographic biases in liver disease diagnosis using the provided gender and age data.

Strengths

• Contains 583 total patient records with a clear class label for liver disease.
• Includes 10 clinical and demographic features per patient, such as bilirubin levels and protein counts.
• Explicitly documents the handling of age data, with patients over 89 listed as age 90.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Last update date is unknown; freshness unverified.
• Data may reflect geographic bias inherent to its collection from a single region in India.

Provenance

Source

UCI Machine Learning Repository (Lichman, M., 2013).

Collection Method

Collected patient records from the North East of Andhra Pradesh, India.

Geography

North East of Andhra Pradesh, India.